This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: -fsquangle
- To: egcs-bugs at cygnus dot com
- Subject: Re: -fsquangle
- From: papadopo at shfj dot cea dot fr (Dimitri PAPADOPOULOS-ORFANOS)
- Date: Wed, 25 Nov 1998 09:31:58 +0100
- Cc: martin at mira dot isdn dot cs dot tu-berlin dot de
> > 1) `libvidaIOfast.so' is a C library compiled and linked with egcs-1.0.2.
> > What does `.stabstr: malformed string table, initial or final byte' mean?
>
> From your description, this sounds like a bug in Sun as: The assembler
> should not produce object files that the linker refuses. Please report
> it as a bug to Sun.
I don't know.
I have one version of egcs-1.1b with -fsquangle enabled and one version
without -fsquangle. I also have some dynamic libraries and programs which
call these libraries.
I tried to build all C++ libraries and programs with -fsquangle, and
then got all these error messages at link-time.
Then I built the same libraries and programs with the -fsquangle-less
version, and everything works fine!
`libvidaIOfast.so' is a C library and has remained unchaged for the
last few months.
> Technically, the message is quite clear: There is a string in the
> .stab section which the linker does not like. You'd have to consult
> your ld documentation for an exact meaning of the message.
>
> > 2) I realize that `-fsquangle' is still experimental.
> > How to give feedback on `-fsquangle'?
>
> As usual, report bugs to egcs-bugs. I'd analyse all object files with
> /usr/ccs/bin/nm, and compare the symbols in the mangled form, instead
> of the pretty form.
Does this mean I should send compiled object files to the list?
I cannot isolate the -fsquangle bugs in small source files.